Issues

• Block‑AP Training Performance on H100

  OpenGVLab/EfficientQAT#34 · Filter by author bimalmagar10 opened on Jan 19, 2026

• About training setting of Llama3-8B

  OpenGVLab/EfficientQAT#33 · Filter by author HXuan-Wang opened on Dec 24, 2025

• about the training data format of e2e training

  OpenGVLab/EfficientQAT#32 · Filter by author brisker opened on Dec 2, 2025

• In main_e2e_qp.py, the default value of max_steps must be set to -1!

  OpenGVLab/EfficientQAT#31 · Filter by author efsotr opened on Nov 12, 2025

• About Dataset

  OpenGVLab/EfficientQAT#30 · Filter by author nostalgiaa-a opened on Aug 14, 2025

• How are the quantization parameters of block_ap passed to the e2e-qp stage?

  OpenGVLab/EfficientQAT#29 · Filter by author tayton42 opened on May 26, 2025
  4

• Release artifacts (models, dataset) on Hugging Face

  OpenGVLab/EfficientQAT#28 · Filter by author NielsRogge opened on May 20, 2025
  1

• colab t4

  OpenGVLab/EfficientQAT#27 · Filter by author werruww opened on Feb 24, 2025

• 其它模型是否支持？

  OpenGVLab/EfficientQAT#26 · Filter by author Code-KingTT opened on Dec 3, 2024
  1

• Llava 1.6 Instruction Finetuning

  OpenGVLab/EfficientQAT#25 · Filter by author gdsaikrishna opened on Nov 28, 2024
  2

• Is 7B llama speed expected to be slow?

  OpenGVLab/EfficientQAT#19 · Filter by author w32zhong opened on Aug 30, 2024
  2

• why model output is too much slow?

  OpenGVLab/EfficientQAT#18 · Filter by author dahwin opened on Aug 27, 2024
  1

• GGUF

  OpenGVLab/EfficientQAT#3 · Filter by author maxim-saplin opened on Jul 26, 2024
  8